Feature Request : MPE support

[Echoes in the Attic]

which seems to be the appropriate behavior (of working mono)

Well, you obviously haven't understood how our concept works and you implied it wasn't implemented "properly". It's mono per channel, which is an extension of that "just mono" of the synths you mention. They seem to do it right because they have the minimum implementation of MPE whereas we strive for a superset.

Cool man, keep insulting me when i was trying to help explain some nuances of different mpe behaviors. If you would cool down and stop being such an asshole for no reason, you'd come to your senses and read what i wrote carefully. I started with "Before you go ahead and properly implement MPE...", so I acknowledged that you have not yet supported it fully so how could I be criticizing your implementation of MPE when you haven't implemented it yet?!?! Jesus Christ this is infuriating. I spend the time to explain how some synths have done it right (and a couple havent quite) so that when you get to it, you have that to consider but instead of "hey thanks for the heads up, we'll take that into account if and when we fully implement MPE", I get "you obviously haven't understood how our concept works". Yes that's it, i just don't understand. So therefor my explanation of my observed MPE behavior is irrelevant and i should never have brought it up. I should have perhaps assumed that you are not interested in what anyone else does because we can just go use another product instead if we don't like your "superset" way, which i guess means a crippled version of an implementation? Whereas those other lowly developers just do a boring "minimum implementation". What the hell does that even mean. And how disrespectful to other developers who actually made the effort to do something well.

If you want mono overall, choose 1 voice polyphony. Simple.

Wow. Just wow. You haven't heard a damn thing I've said. When in a mono mode, your synths still play polyphonically when receiving on multiple channels. I'm not going to go over the reasons again why it's beneficial to allow mono while still receiving separate expression channels. I tried to explain it but you tried to make me feel like I was an idiot for bringing it to your attention and that it "had nothing to do with MPE" despite the fact that i am describing how officially MPE supported synths behave. Everyone else is wrong and you are right. Stupid me.

In over a decade on this forum the only developer I've found this disrespectful and arrogant is Markus from Tone2. Congrats on joining an elite club.

[Urs]

So, switching, say, Diva, to 1 voice polyphony doesn't do exactly what you describe? It does, doesn't it?

If anything I'm sorry that this topic is currently so far away from me, I hadn't had the solution right there. There was no need to lecture/ridicule me with some random synths from the competition. Things like that put me on the edge when I have other kinds of stress. For a moment there I was really frustrated that MPE wasn't possible with our concepts, based on false information.

In the after thought, MPE and u-he will be just fine. That esoteric thing you describe will be just fine with our stuff. None of our concepts collide with MPE. Good. Makes me happy.

[Urs]

Given this present state of affairs my recommendation to software synth makers who are interested in this stuff and have already implemented very nice polyphonic multi-channel stuff in their products, is not to worry too much about trying to support every last detail of the spec at the moment.

That's exactly where we are. The multichannel MIDI support was the first major step. ACE, Bazille, Diva and Hive work very well with most MPE controllers out there. That's because most of these controllers support more than just MPE, they're usually also happy with what used to be called "Guitar Mode".

Repro-1 is our first synth to support wide Pitchbend ranges (0-24, 36, 48), even though it isn't MPE copatible at all. However, that code, however small a fragment it is, will join our other synths.

We have an internal ticket to replace our two fixed MIDI Controller assignments (Breath, Expression) with a set of 2 or 4 MIDI Controllers of choice. This however involves quite a bit of refactoring and testing, so we haven't done it yet.

What's the missing from full MPE support are some of the formal aspects, like Channel 1 for global control, reacting on MPE on/off messages and stuff. I'm not sure how deeply these have been implemented in the controllers out there yet.

So I don't think we're far off. Afaik the Linnstrument and Rise work just fine with what we got.

[thetechnobear]

I use an Eigenharp and Soundplane with all the U-He synths, and yes, they are also fine in 'guitar mode'

Im not too fussed about the NRPNs to switch to MPE mode, or change the PB range - which although controllers do implement, I dont think any software synth does (Strobe2, Madrona labs, Falcon - do not) , perhaps equator does but its a private Roli only synth, so frankly doesn't count


but as I pointed out before (a year ago?!) if we had CC74 and PB range of 48, then we would not have to switch out of MPE mode to guitar mode for u-he synths, it would still work. it would give us 95% of what we need

(it should also be noted that BWS's MPE support will not work in guitar mode, since it is hard-coded to send CC74 for the Y axis, so to get around this, we would have to use a host within a host, something like Bidule to map CC74 to CC1... which is a huge kludge since you lose automation etc)

Yes global control on Ch1 would be nice, particularly for breath/ribbon/pedal input, but thats the 'icing on the cake'.

I guessed Id hope a quick fix would be possible for just adding CC74 to expression, and extending the PB range, a pragmatic solution... rather than wait for the perfect solution

[Urs]

but as I pointed out before (a year ago?!) if we had CC74 and PB range of 48, then we would not have to switch out of MPE mode to guitar mode for u-he synths, it would still work. it would give us 95% of what we need

We'll get there. I hadn't had the strength to do it for the rev 3898 maintenance update, nor the Diva NKS.

But we will shell out some updates soon, at least the PB range should be in for it.

[NothanUmber]

but as I pointed out before (a year ago?!) if we had CC74 and PB range of 48, then we would not have to switch out of MPE mode to guitar mode for u-he synths, it would still work. it would give us 95% of what we need

We'll get there. I hadn't had the strength to do it for the rev 3898 maintenance update, nor the Diva NKS.

But we will shell out some updates soon, at least the PB range should be in for it.

Thanks for adding MPE support step by step (well, the parts that practically count ), u-he stuff rocks! Extended PB range and CC74 support will be really helpful, the rest is more "icing on the cake" as technobear wrote.
Hope you have a lot of fun with your Linnstrument? (if yes then perhaps Zebra 3 will become the ultimate poly expression synth? Well at least it is nice to dream )

[Stollmeister]

I joined the Roli Seaboard crowd. Pitch bend 48 and cc 74 for Zebra would be super!

[NothanUmber]

I joined the Roli Seaboard crowd. Pitch bend 48 and cc 74 for Zebra would be super!

Zebra (2) would need voice per channel support (pitch bend, modulation etc.) first - which as far as I understood Urs would be difficult to add to the existing application. But let's keep fingers crossed for Zebra 3

The other u-he synths (Diva, Bazille, I think also Hive) are already 99% there though (they have voice per channel support for pitch bend, modulation and expression). So there the increased pitch bend range and cc 74 would be a relatively small step for u-he, but a big step for (poly expression fanclub) mankind

[Stollmeister]

I joined the Roli Seaboard crowd. Pitch bend 48 and cc 74 for Zebra would be super!

Zebra (2) would need voice per channel support (pitch bend, modulation etc.) first - which as far as I understood Urs would be difficult to add to the existing application. But let's keep fingers crossed for Zebra 3

The other u-he synths (Diva, Bazille, I think also Hive) are already 99% there though (they have voice per channel support for pitch bend, modulation and expression). So there the increased pitch bend range and cc 74 would be a relatively small step for u-he, but a big step for (poly expression fanclub) mankind

Good point! However, I have found some great monophonic presets for Zebra. For them, having the increased pitch bend and cc74 would be great.

[Howard]

...I have found some great monophonic presets for Zebra. For them, having the increased pitch bend and cc74 would be great.

You could tweak those presets for 48-semi pitchbend (and MIDI-learn CC74 to a Y-knob of an X/Y pad? But I'm not 100% sure what that's about...)

[hyperscientist]

I'm just reporting the way the other fully supported MPE synths behave in MPE mode.

Thanks for your research, it was an interesting read actually. I myself used to try my Seaboard with all synths I own (including u-he) with some custom fixes done with max for live, but I got tired with it quickly and I'm waiting for better times using only Equator for MPE duties in the meantime. It's good to have more persistent bleeding edge riders like yourself out there - vi vendors and everyone else should be thankful

[Restless]

I bookmarked this thread since I got a Roli Seaboard 25 a month or so ago. I'm on the lookout for nice soft-synths and u-he ones are on the top of my list if MPE is supported. So I have to say I was gutted to see your negativity towards this spec. A spec being in draft should not be seen as something negative, especially with standards that are slow to evolve like midi - the fact its even part of a draft spec is almost a miracle and great progress compared to, for example, the state of play when I tried to use an Eigenharp last decade.

As a developer (albeit not of music stuff) I can understand why this stuff may cause frustrations at times, especially when it is something of a small niche right now. But you could look at it as an opportunity and try to embrace it more convincingly. I'd like to become a customer of yours, and even maybe help untangle some of the confusion when I'm slightly more up to speed on the spec and its implications in different situations, but not if you dont really want to embrace MPE.

After having bought every u-he synth minus Hive, I agree that URS attitude towards this whole issue is really putting me off. At the moment I'm focusing on learning the technique on the seaboard rise, but until this changes, no more U-he (or any other non MPE ayntha for me) synths for me. I obviously passed on repro-1, and am waiting on my free Zebra 3, but after experiencing what's achievable with MPE, even if only with the bland sounding Equator, I can't see myself spending more money on synths that don't fully implement it.

And this standard, especially thanks to Roli having making affordable controllers, is, in my opinion, going to pick up. A shame really.

[Stollmeister]

I bookmarked this thread since I got a Roli Seaboard 25 a month or so ago. I'm on the lookout for nice soft-synths and u-he ones are on the top of my list if MPE is supported. So I have to say I was gutted to see your negativity towards this spec. A spec being in draft should not be seen as something negative, especially with standards that are slow to evolve like midi - the fact its even part of a draft spec is almost a miracle and great progress compared to, for example, the state of play when I tried to use an Eigenharp last decade.

As a developer (albeit not of music stuff) I can understand why this stuff may cause frustrations at times, especially when it is something of a small niche right now. But you could look at it as an opportunity and try to embrace it more convincingly. I'd like to become a customer of yours, and even maybe help untangle some of the confusion when I'm slightly more up to speed on the spec and its implications in different situations, but not if you dont really want to embrace MPE.

After having bought every u-he synth minus Hive, I agree that URS attitude towards this whole issue is really putting me off. At the moment I'm focusing on learning the technique on the seaboard rise, but until this changes, no more U-he (or any other non MPE ayntha for me) synths for me. I obviously passed on repro-1, and am waiting on my free Zebra 3, but after experiencing what's achievable with MPE, even if only with the bland sounding Equator, I can't see myself spending more money on synths that don't fully implement it.

And this standard, especially thanks to Roli having making affordable controllers, is, in my opinion, going to pick up. A shame really.

A pity you missed out on Repro. It is a joy to play with the Roli. The pressure and the pitchbends work great.

You find Equator bland? I haven't A/B tested it with Zebra, but so far I really enjoy playing it. I have used Equator for polyphonic parts and Repro and Zebra for monophonic stuff with the Roli.

[EvilDragon]

MPE is far from being a standard since apparently they cannot agree on all the details to finalize the v1.0 spec, so it's still a draft specification rather than a full-blown standard. I understand Urs' point of view.

[Urs]

Thing is, if a small company like us keeps implementing new standards as they come and change, we'll be broke before we finish anything. I think the past 5 years we've spent more than 20% of our time on whatever standard had to be implemented, but I doubt that the revenue was all that. So it's sheer self defense if we put these things aside until we can sure that we'll survive the effort. That also means we can't just do something with a draft or a beta standard anymore. We used that up on RE, NKS, Linux, AAX, VST3 and the current state of MPE support (of which I think NKS was the most profitable). No more, until it's set in stone.